Over the years, I have learned to trust my instincts. And during a pre-study assessment several years ago, if I had not listened to my “gut,” I would have chosen the wrong site for our study.

Based on my initial contact with the site and the feasibility paperwork, it seemed it was an experienced research site that worked with a local physician in trial conduct. I anticipated a routine assessment. Unfortunately, signs of trouble appeared almost immediately.

I arrived at the site a little early (the day of the meeting). The nurse was on the phone and asked if I would wait briefly in the lobby. This was not unexpected. The last moments of preparation are critical. A full 25 minutes elapsed before the nurse retrieved me, only to deposit me in a back conference room before racing to answer another call. It seemed odd, but I put it out of my mind and organized my presentation materials.

Ten minutes later, the nurse entered the room. She explained that the other study nurse had called in sick, and the research assistant was running late. She had to cover the phones until the assistant arrived. My first thought was they had staffing issues. But then I realized the nurse couldn’t help if her colleagues weren’t there. I informed the study nurse that I understood, and we would make the best of it. I would review trial design and eligibility with the PI, and the nurse would join us when the research assistant arrived—an acceptable workaround.

The site feasibility paperwork indicated the investigator/staff had conducted several trials in the study therapeutic indication. But their feasibility data revealed a conflict in patient enrollment projections. They treated 10 to 15 patients a year for the disease under study. They had projected enrolling 50 patients in six months.

The site had a strategic arrangement with a local urgent care, and this was the primary measure for patient recruitment (according to feasibility data). While the numbers were promising, the ratio to patients treated and enrollment projections was not logical. I planned to clarify this with the PI.

The PI was professional and provided ample time for the protocol discussion, but was unable to clarify the site enrollment process. He was surprisingly uninvolved with patient identification/recruitment; the recruitment department was responsible for providing feasibility enrollment data. They also were responsible for patient identification via database searches and targeted calling. The PI could not confirm an urgent care affiliation. He then excused himself to attend his patients and advised me to confirm information with the nurse. But she had not been able to join the meeting due to her missing colleagues.

After 20 more minutes of waiting, I became concerned. I had about 90 minutes until I had to leave. If I did not speak to the study nurse, I would run out of time to complete the remaining assessments, which included the site experience, recruitment discussion and facilities tour.

I opened the conference room door to find the study nurse engaged in conversation with an individual I presumed to be the research assistant. I asked the research nurse when we would be able to continue our discussion. She apologized and followed me back into the conference room. She said they were resolving an unexpected study issue; because the other study nurse was working at their sister site, they had limited resources to cover multiple responsibilities.

A couple of questions occurred to me: First, didn’t the study nurse originally say her colleague was ill; second, did they have more study responsibilities than staff required to fulfill them?

We began to discuss the patient recruitment process. I asked the study nurse to clarify the discrepancy with enrollment projections. Without hesitation, she said central recruitment was responsible for patient enrollment data, and I would need to contact that department to clarify. She was unable to confirm any relationship with the local urgent care facility.

I wondered: Why were the PI and study nurse unable to discuss enrollment projections, and had inaccurate feasibility data been provided?

At that point, I realized the unforeseen circumstance was no longer a sufficient explanation for insufficient staffing and data provision. I began to ask targeted questions to validate my concerns regarding staff retention, study workload, and therapeutic experience. I confirmed three more suspicions: The staff had worked there less than a year; the site was running 19 trials, with only two study nurses and one research assistant; and not one staff member had experience in the study therapeutic indication, contrary to feasibility data.

I finished the assessment activities with the study nurse, but had decided not to recommend the site. Ideal investigational sites are transparent with experience and capabilities, are engaged with patient recruitment, and have sufficient research staff to manage trial responsibilities. Lacking these core elements can result in protocol deviations and GCP noncompliance. 

Elizabeth Blair Weeks-Rowe, LVN, CCRA, has served in a variety of clinical research roles including CRA, CRA trainer, CRA manager and clinical research writer. She works in relationship development/study startup in the CRO industry. 
Email ebwcra@yahoo.com.

This article was reprinted from CWWeekly, a leading clinical research industry newsletter providing expanded analysis on breaking news, study leads, trial results and more. Subscribe »